CASE BASED REASONING UNTUK MENDIAGNOSIS PENYAKIT MATA (Studi kasus : RSUD Kab. Pangkep Kec. Pangkajenne Provinsi Sulawesi Selatan )
FADLI TAMRIN, Drs. Retantyo Wardoyo, M.Sc., Ph.D
2016 | Tesis | S2 Ilmu KomputerSebagai salah satu dari lima indera manusia, mata berguna untuk menyampaikan informasi visual kepada otak (Faizal, 2012). Mata yang mengalami gangguan fungsi yang disebabkan penyakit, maka informasi yang dihasilkan dari lima indera akan mengalami ketidakseimbangan yang sudah pasti berujung pada gangguan dalam aktivitas keseharian manusia. Model yang dibangun dalam penelitian ini adalah model untuk mendiagnosa penyakit mata menggunakan Case Based Reasoning (CBR) dan Certainty Factor (CF). CBR dalam penelitian ini digunakan untuk mencari kasus-kasus penyakit mata sebelumnya yang tersimpan pada basis kasus yang memiliki tingkat kemiripan yang paling mirip dengan problem baru. Sedangkan CF merupakan metode yang digunakan untuk mencari nilai tingkat keyakinan terhadap penyakit mata yang dihasilkan oleh proses CBR. Hasil penelitian menunjukkan bahwa sistem mengenali 28 data yang sesuai dengan data riil hasil diagnosa pakar dan 2 data yang tidak sesuai dengan data riil hasil diagnosa pakar atau bisa dikatakan nilai akurasi sistem sebesar 93%.
As one of the five human senses, the eye is useful to convey visual information to the brain (Faizal, 2012). If the eyes have impaired function caused by a disease, then the information that was generated by the five senses will be disrupted; therefore it will certainly lead to an interruption in daily activities. To overcome the problems of eye diseases we create an expert system. The model that is created in this research is an expert model diagnostic of eye diseases using Case Based Reasoning and Certainty Factor. Case Based Reasoning (CBR) is a reasoning system of computers that uses old knowledge to solve new problems. CBR provide solutions for new problems by looking at old cases that closely resembles the new problem. The process of calculating the value of similarity between the new problem with the base case uses the method of nearest neighbor similarity, after obtaining similar base cases with the new problem, we calculate the level of confidence. Certainty factor (CF) is a method to prove if a fact is definite or non-definite in a form of metric, cases on base case that are similar with the new problem and exceeds the threshold value of 0.6 will be used to found the case with the greatest value of the level of confidence, and that case with the highest level of confidence will be used as a solution to solve the new problem. The results showed that the system recognizes 28 data corresponding to the real data expert diagnosis and 2 data that does not correspond to the real data or expert diagnosis can be said value system accuracy by 93 % .
Kata Kunci : Kata kunci : Case Based Reasoning , Certainty Factor